Checkpoint 1: EDA

By Arturo, Payson, and Victoria

Does healthcare access affect the number of preventable hospital stays of certain racial groups at the county level?

Why this matters:

• Reduce preventable hospitalizations
• Equitable resource allocation
• Focus efforts where help is needed most

<<<<<<< HEAD ## Why It Matters:

======= >>>>>>> 7ee0a101efb53b4520a6b63566b9fd4e4ad4fa0e ————————————————————————

National Data

Source:
County Health Rankings 2025

<<<<<<< HEAD Variables of Interest:
- Preventable Hospital Stays, Primary Care Physician Ratio, Uninsured Rate - others things - County-level racial groupings (White Majority vs White Minority)

Preprocessing Highlights:
- explain how we did white majority vs minority ======= Raw Data:

• County-level health outcomes(ex)
• Healthcare access variables(ex)
• Disaggregated by race

Cleaned Data:

• Filtered and cleaned to complete observations
• Created “White Majority” vs. “Minority” indicator >>>>>>> 7ee0a101efb53b4520a6b63566b9fd4e4ad4fa0e

Preventable Hospitalizations by County

Regression Analysis of Racial Disparities in Preventable Hospital Stays

Planned Approach:
- Linear regression with interaction terms
- Outcome: Preventable hospital stays
- Predictors: Uninsured rate, provider ratio, race group

Modeling Strategy:
<<<<<<< HEAD - Use ….

Justification:
- give reasons ======= - Explore variable selection and regularization >>>>>>> 7ee0a101efb53b4520a6b63566b9fd4e4ad4fa0e

Moving into modeling and interpretation

<<<<<<< HEAD ** Completed:**
- Defined research question - Cleaned and preprocessed data - Created two EDA visualizations ======= Completed:
- Defined research question
- Cleaned and preprocessed dataset
- Created initial EDA visualizations >>>>>>> 7ee0a101efb53b4520a6b63566b9fd4e4ad4fa0e

Next Steps:
- Fit regression models with race × access interactions
- Interpret disparities and validate model - Create visual summaries and poster